Robots that assist elderly or disabled persons, or even anyone in their day-to-day tasks, can lead to a huge improvement in quality of life. At ROCON we are pursuing domestic mobile manipulators, as well as UAVs for monitoring the persons. Our next goal is to integrate these two platforms into an overall framework that will both monitor the persons and assist them on the ground.
The research projects below are currently open. We usually have MSc, BSc, and PhD positions available. If you are interested in a project, check its detailed description and contact the people listed there.
If you would like to work with us on something else, tell us about it! We are always looking for new ideas and motivated people with whom to work on them.
Robot localization in 3D space is an essential task for many applications in this research field. One of the possible ways to achieve this is to use a camera network, i.e. a set of interconnected cameras which can precisely detect and track the position of an object in the 3D space.
A wide variety of UAV tasks involve detecting and tracking an object, and then controlling a desired UAV trajectory relative to this object. Examples include following a person, video inspection of an object in the environment, etc. This project is for a team of at least two students, one working on detection, and the other on control.
This project concerns controller design for the Cyton Gamma 7DOF robot arm available in our group. This being a mechanical system, modeling will commence from the Euler-Lagrange equations, leading to a TS model in descriptor form. For the sake of simplicity of the controller implementation, this form will be maintained for the design.
In the context of the development of a new strategy for helping persons with limited mobiliy in wheelchair, we wish to estimate the forces at the shoulder of a person during his moving in wheelchair. To achieve this goal, an observer has to be designed from a mechanical model of the upper body of the person and from experimentations on a wheelchair.
This project will develop observers and controllers for one of the experimental setups in our group. Options include the Quanser rotational inverted pendulum, controlling an Inteco3D crane to move the load along a designated trajectory, etc.
Fuzzy-polynomial approaches have gained considerable interest in the last years for control of nonlinear systems. The stability and design conditions for such models are derived in the form of sum-of-squares, which can be solved using available tools.
Optimistic planning is a recent approach to the predictive control of nonlinear systems, which optimistically explores the space of action sequences from the current state. Due to generality in the dynamics and objective functions that it can address, it has a wide range of potential applications to problems in nonlinear control.
The Sumo Robot contest is an annual event at our faculty. Each year more and more students are interested and join in. In this project, we will start from existing sumo robot hardware (preferably the student's robot!), with the goal of developing better software, so that it could beat more expensive robots by simply being a "smarter" fighter.
An over-the-air system update process is required for many applications. A robust mechanism is recommended to update system firmware remotely to fix bugs and make important software updates. The project goal is to implement a software module for ARM based microcontrollers used in robotic applications to change the application logic.
This project has its main goal the mapping and localization of a mobile robot from the National Instruments in a flat 2D environment. Standard techniques are envisaged in order to create a 2D environment map and to use this in further steps for localization.
We intend to develop a new communication profile for robotic applications. This profile will include standard interfaces and definition of devices to allow communication between different types of robot controllers. The profile will define the communication logic bewteen different controllers and will run on the higher levels of the controller software stack.
This project relates two important applications domains: the perception and control in the field of humanoid robotics. The aim of the project is to develop a stable and reliable human-robot interaction system using cameras (either 2D or 3D) and to use this to control a mobile robot such as a dual arm robot.
This project makes use of ground and/or aerial vehicle to localize this in an outdoor environment. The main aim of the project is to fuse the information from different sensors such as IMU, GPS, visual odometry and get the best localization of the mobile agent.
The challenging problem of fusing the data from central cameras such as 3D depth, infrared, omnidirectional or plain perspective camera is a must in the recent mobile robotics perception systems. Also the active perception task, i.e. the ability of getting the most information from a scene is getting into focus, thus this project aims to tackle this novel problem in the near future.